Neural style transfer (NST)
Neural Style Transfer (NST) is a technique in artificial intelligence that combines the content of one image with the style of another, creating a unique artwork.
Neural networks
MXNet
MXNet is an open-source deep learning framework designed for flexible and efficient training and deployment of neural networks. It supports multiple programming languages and is known for its scalability across multiple GPUs and distributed computing environments.
Adaptive instance normalization (AdaIN)
Adaptive instance normalization (AdaIN) is a technique primarily used in style transfer and other applications in deep learning, allowing for the adjustment of feature statistics.
Motion planning with neural networks
Motion planning with neural networks involves using artificial neural networks to compute feasible paths for autonomous agents or robots within an environment. This approach leverages machine learning techniques to improve the efficiency and adaptability of traditional motion planning algorithms.
Backpropagation
Backpropagation is a fundamental algorithm used in training artificial neural networks by efficiently computing gradients needed for optimization. It enables the adjustment of network weights through the chain rule of calculus, facilitating learning in multi-layer networks.
Neural Turing machine
A Neural Turing Machine (NTM) merges neural networks with external memory resources, enabling the model to learn and store information dynamically.
Multi-task learning
Multi-task learning is a machine learning paradigm that simultaneously learns multiple tasks, sharing knowledge across them to improve performance.
Variational autoencoder (VAE)
Variational autoencoders (VAEs) are generative models that leverage neural networks to learn complex data distributions, enabling effective data generation and representation.
Recurrent neural network
Recurrent neural networks (RNNs) are a class of artificial neural networks designed for processing sequential data, enabling them to maintain a memory of previous inputs.
WaveGAN (GAN for raw audio)
WaveGAN is a type of generative adversarial network (GAN) designed specifically for generating raw audio waveforms. It enables the synthesis of audio signals directly in the time domain without relying on spectrogram or frequency domain representations.